u2k15: krw@ on msdosfs fixes, installer cleanup, GPT fdisk, softraid, dhclient

After a roundabout trip via Warsaw and Zurich I arrived in Berlin late on the 22nd, made a brief appearance at the hackroom to ensure I still knew the way (beware the non-euclidean nature of roads in Berlin!), and then slept the sleep of the traveller to be fresh for hacking.

With the network hackers successfully isolated (bar sporadic Hummpa outbreaks) in a smaller room the 'front room' proved to be a productive environment. Occasionally disturbed by vehement arguments on the theme of the hackathon, utf8 support.

I started off by working on a msdosfs bug (renaming a file in the root directory failed) reported by matthieu@. By the time I had repaired some eye searing whitespace and enabled msdosfs to compile with MSDOSFS_DEBUG so I could debug the issue better, a patch had been submitted to bugs@ by Serguey Parkhomovsky. Which I committed. Job well done!

I then cleaned up some historical baggage in a few install.md files (amd64, armish, i386, landisk, macppc, socppc, zaurus) by replacing some here-document scripting of fdisk with the equivalent 'fdisk -i'.

I nuked some disklabel(8) '-b' left-overs by removing support for the disktab(5) 'b0' and 'b1' capabilities, and nuking the d_packname union with boot file pointers from the disklabel itself. And tweaking the manpages appropriately of course.

I spent some time helping stsp@ fix installboot(8) to handle softraid crypto installs, while stoutly maintaining my ignorance of softraid crypto. I failed to see a bug stsp@ introduced, which lead to some hilarity a few days later when I backed out the change at the same time that stsp@ was committing a real fix. Fortunately stsp@ was able to set things right, and installboot(8) now works for softraid crypto setups and non-softraid setups.

Next, I finally committed GPT editing capability to fdisk(8). While aesthetic interaction with the user remains an issue, at least GPT partitioning can now be seen and manipulated without outside tools. The aesthetics will improve! While tweaking the man page I managed to cause mandoc to seg fault. Ingo was appalled and immediately fixed it. Once I remembered to tell him.

Lastly I committed 'domain search' support in dhclient(8) after prodding from Claudio. Which of course means there is now a crying need to add RFC1035 parsing in the dhcpd/dhclient config files. Sigh.

Overall another excellent hackathon. Thanks to stsp@ and the other Berlinians for organizing it and thanks to IN-Berlin for hosting it.

Thanks for the report and the awesome work, Ken!